REVIEW SUMMARY◥

CHEMICAL BIOLOGY

Chemically induced proximityin biology and medicineBenjamin Z. Stanton,* Emma J. Chory,* Gerald R. Crabtree†

BACKGROUND: Nature has evolved elegantmechanisms to regulate the physical distancebetween molecules, or proximity, for a widevariety of purposes. Whether it is activation ofcell-membrane receptors, neuronal transmis-sion across the synapse, or quorum sensingin bacterial biofilms, proximity is a ubiquitousregulatorymechanism in biology. Over the pasttwo decades, chemically induced proximity hasrevealed that many essential features and pro-cesses, including protein structure, chromosomalarchitecture, chromatin accessibility, transcript-ion, and cellular signaling, are governed by theproximity of molecules. We review the criticaladvances inchemical inducersofproximity (CIPs),

which have informed active areas of research inbiology ranging from basic advances to the de-velopment of cellular andmolecular therapeutics.

ADVANCES: Until the 1990s, it was unclearwhether proximity was sufficient to initiatesignaling events or drive their effect on tran-scription. Synthetic small molecule–induceddimerization of the T cell receptor providedthe first evidence that proximity could be usedto understand signal transduction. A distin-guishing feature of small-molecule induced-proximity systems (compared to canonicalknockdown or knockout methods) is the abil-ity to initiate a process midway and discern the

ensuing order of events with precise temporalcontrol. The rapid reversibility of induced prox-imity has enabled precise analysis of cellularand epigenetic memory and enabled the con-struction of synthetic regulatory circuits. Inte-gration of CRISPR-Cas technologies into CIPstrategies has broadened the scope of thesetechniques to study gene regulation on timescales of minutes, at any locus, in any geneticcontext. Furthermore, CIPs have been used to

dissect the mechanismsgoverning seeminglywell-understood processes, rang-ing from transport ofproteins between theGolgiand endoplasmic retic-ulum to synaptic vesicle

transmission. Recent advances in proximity-induced apoptosis, inhibition of aggregation,and selective degradation of endogenous pro-teins will likely yield new classes of drugs inthe near future.

OUTLOOK:We review fundamental concep-tual advances enabled by synthetic proximityas well as emerging CIP-based therapeutic ap-proaches. Gene therapy with precise regulationand fully humanized systems are now possible.Integration of proximity-based apoptosis throughcaspase activation with chimeric antigen re-ceptor (CAR) T cell therapies provides a safetyswitch, enabling mitigation of complicationsfrom engineered immune cells, such as graft-versus-host disease and B cell aplasia. Further-more, this integration facilitates the potentialfor repopulation of a patient’s cells after suc-cessful transplantation.With the recent approv-al of CTL019, a CAR T cell therapeutic fromNovartis, integrated strategies involving theuse of CIP-based safety switches are emerg-ing. Innovative exemplars include BPX-601(NCT02744287) and BPX-701 (NCT02743611),which are now in phase 1 clinical trials. By usinga similar proximity-based approach, condition-al small-molecule protein degraders are alsoexpected to have broad clinical utility. Thisapproach uses bifunctional small moleculesto degrade pathogenic proteins by dimerizingwith E3 ubiquitin ligases. Degradation-by-dimerization strategies are particularly ground-breaking, because they afford the ability torepurpose any chemical probe that bindstightly with its pathogenic protein but whichmay not have previously provided a directtherapeutic effect. We anticipate that thetranslation of CIP methodology through bothhumanized gene therapies and degradation-by-dimerization approaches will have far-reaching clinical impact.▪

RESEARCH

Stanton et al., Science 359, 1117 (2018) 9 March 2018 1 of 1

The list of author affiliations is available in the full article online.*These authors contributed equally to this work.†Corresponding author. Email: crabtree@stanford.eduCite this article as B. Z. Stanton et al., Science 359,eaao5902 (2018). DOI: 10.1126/science.aao5902

Chemically induced proximity. (Top) Left: Small molecules (hexagons) bind proteins of interest(crescents), dimerizing them to increase the effectivemolarity of reactions. [A] monomeric protein and[AB*] dimer concentrations; arrows, position coordinates. Middle: Synthetic dimerizers tag proteins(blue circles) for proteasomal degradation (red rods). Right: Homodimerizing molecules form killswitches for apoptosis. (Bottom) CIPs mimic cellular processes. Left: Protein transport mechanisms—nuclear import and export, membrane fusion, and protein folding. Middle: Regulation of gene activationby binding to DNA or chromatin (spheres with white strands), through recruitment of transcriptionalactivators or repressors (blue and red arrows). Right: Signal transduction pathways.

ON OUR WEBSITE◥

Read the full articleat http://dx.doi.org/10.1126/science.aao5902..................................................

on October 1, 2020

http://science.sciencem

ag.org/D

ownloaded from

REVIEW◥

CHEMICAL BIOLOGY

Chemically induced proximityin biology and medicineBenjamin Z. Stanton,1,2* Emma J. Chory,1,3* Gerald R. Crabtree1,4†

Proximity, or the physical closeness of molecules, is a pervasive regulatory mechanism inbiology. For example, most posttranslational modifications such as phosphorylation,methylation, and acetylation promote proximity of molecules to play deterministic roles incellular processes.To understand the role of proximity in biologic mechanisms, chemicalinducers of proximity (CIPs) were developed to synthetically model biologically regulatedrecruitment. Chemically induced proximity allows for precise temporal control oftranscription, signaling cascades, chromatin regulation, protein folding, localization, anddegradation, as well as a host of other biologic processes. A systematic analysis of CIPs inbasic research, coupled with recent technological advances utilizing CRISPR, distinguishesroles of causality from coincidence and allows for mathematical modeling in syntheticbiology. Recently, induced proximity has provided new avenues of gene therapy andemerging advances in cancer treatment.

Biochemical processes are often regulatedby the physical distance, or proximity, be-tween molecules to initiate an effect. Prox-imity plays both a ubiquitous and essentialrole in biology, whether it relates to indi-

vidual cells, as in confining enzymes within dense-ly packed organelles, or whole populations, aswith quorum sensing in bacteria. The importanceof utilizing small molecules to induce proximityof proteins was recognized upon the discoverythat the Src homology 2 (SH2) domain of tyro-sine kinases mediates signal transduction bybinding phosphotyrosine in the absence of ca-talysis (1). Later research showed that acetylation,methylation, ubiquitination, and a host of othertransient or stable protein modifications recruitproteins that influence many processes, such asgene regulation and protein degradation. Therealization that these changes in localizationcould produce distinct cell-fate decisions led toa fundamental question, “How does a quanti-tative change in localization produce discretebiologic responses?” The answer appears to liein the simple fact that the probability of aneffective collision between two molecules is athird-order function of distance (2). This simplerelation allows steep concentration gradients toproduce qualitative changes, such as cell lineagecommitment. Yet, mechanisms other than prox-imity, like allostery, might mediate these biologic

responses. How does one separate the consequen-ces of these processes?The effects of proximitywere first distinguished

from allosteric or alternative effects by the syn-thesis of a bivalent molecule, FK1012, that boundits ligands with no detectable allosteric changes.The nontoxic molecule simultaneously binds twoFK binding proteins (FKBPs), each of which is a108–amino acid prolyl isomerase. FK1012 was firstused to homodimerize the intracellular domain ofthe T cell receptor (TCR) zeta chain (Fig. 1), pro-ducing signaling events that reproduce trans-membrane signaling by the TCR (3). This firstdemonstration that chemically induced proxim-ity (also referred to as chemically induced dimeri-zation) could activate signaling was followed bysimilar approaches with Ras signaling (4), deathreceptor signaling (5), and transcriptional activa-tion (6), among others. Each case supported acausative role of simple proximity in qualitativecellular changes. Although the role of proximityin the absence of allostery is still debated (7),we will focus this review on the emerging use ofinduced proximity with small molecules in re-solving complex biologic questions and designingnew therapeutic strategies.

Tool kits to explore proximity in biology

The first chemical inducer of proximity (CIP),FK1012 (3), a homodimer of FK506, was followedby many others (Fig. 1). These molecules havethe common feature of binding two peptide tagson either side of each molecule. Given that in-duced proximity is observed within minutes, onecan study the immediate, primary effects of ac-tivating a specific molecule without concern fordelayed toxic effects of the dimerizer on prolifera-tion, transcription, or other much slower processes.Often these molecules are naturally occurring andillustrate how biology regulates proximity to its

own benefit. For example, FK506 binds FKBP12on one of its sides and calcineurin, a phospha-tase essential for immunologic activation, on theother side (Fig. 1), illustrating how induced prox-imity and inhibition of calcineurin by FKBP12functions in immunosuppression. Other examplesof naturally induced proximity include cyclospo-rine A (Fig. 1), which also inhibits calcineurin byrecruiting cyclophilin to its active site (8), therebyinhibiting phosphatase activity and nuclear local-ization of NFATc (nuclear factor of activated T cells)family members (9). Rapamycin, an immunosup-pressant structurally related to FK506, also sharesthe primary target FKBP and acts through for-mation of a ternary complex (Fig. 1) with theFKBP12-rapamycin-binding (FRB) domain of tar-get of rapamycin (TOR) kinases (10, 11).In plants, induced proximity with abscisic acid

(Fig. 1) blocks germination and also inducesleaves to abscise in the fall. It functions by induc-ing proximity of the monomeric receptor Pyl tothe protein phosphatase ABI1 (12, 13). This mol-ecule is present at high concentrations in our dietsand is not toxic in humans. Similarly, gibberellin(Fig. 1), which promotes germination and stemelongation in plants, functions by induced prox-imity of the receptor GID1 and hormone GA1 (14).

Dynamics of chemically induced proximity

Over the past 20 years, CIP technology has ad-vanced from its origins to afford methods to un-derstand signaling, transcription, and proteinlocalization on rapid time scales. Much of theprogress hinges on the ability to initiate biologicprocessesmidpathway in vivo, such as downstreamof a signal-activation event, and then discern theorder of reactions after induced activation. Thepower of this approach arises from the fact thattemporally ordering events places rigorous limitson causality.Paradoxically, the responses from chemically

induced proximity are often more robust thanthose from rigid protein fusions, especially incases where a protein fusion can result in sterichindrances that prevent functionality. Further-more, chemically induced proximity providesminute-by-minute kinetic analysis, allowing pre-cise mathematical modeling. The fundamentalconcept of effective molarity—that a localizedconcentration within solution may differ fromthe bulk concentration—underlies the rationale andpracticality of using chemically inducedproximity tostudy complex biologicmechanisms. Proximity be-comes a critical regulator of cellular processes bythe fact that the probability of an effective inter-action between two molecules is a function of thedistance between them. This phenomenon can beobserved by considering the scaling relationshipsbetweenphysical distance and reactionprobability.In most relevant cases, reaction rate scales withconcentration (the inverse cubed root of particledensity), which scales with mean interparticledistance, i.e., the closeness of molecules (2).The contributions of effectivemolarity are read-

ily observed in natural processes such as proteincompartmentalization within organelles, mem-brane localization, and protein scaffolds.Molecular

RESEARCH

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 1 of 9

1Departments of Pathology and Developmental Biology,Stanford University School of Medicine, Stanford, CA 94305,USA. 2Division of Preclinical Innovation, National Center forAdvancing Translational Sciences, National Institutes ofHealth, Rockville, MD 20850, USA. 3Department of ChemicalEngineering, Stanford University, Stanford, CA 94305, USA.4Howard Hughes Medical Institute, Stanford UniversitySchool of Medicine, Stanford, CA 94305, USA.*These authors contributed equally to this work.†Corresponding author. Email: crabtree@stanford.edu

on October 1, 2020

http://science.sciencem

ag.org/D

ownloaded from

scaffolds increase effective molarity in biochem-ical processes such as transcription, translation,and biosynthetic pathways. Protein scaffolds canenhance the speed of enzymatic reactions by sev-eral thousandfold (15, 16). Organelles sequestercritical reaction components through compart-mentalization to increase the effective molarityof relevant substrates.Previously, mathematical models of reaction-

diffusion systems have been used to describe dy-namic biologic processes that correspond to achange in concentration with respect to space,time, and changing substances (reactions). Someexamples of reaction-diffusion models in biologyinclude those that explain the improved enzymecatalytic efficiency resulting from compartmen-talization (17) and those that describe the im-proved kinetics of push-pull networks in whichtwo enzymes control signal transduction path-ways in an antagonistic manner (18). Similarly,the effective concentration increase at a CIP re-cruitment site can be understood by consideringany dimerization event as a reaction occurring ina classic reaction-diffusion system (Fig. 2A) (be-cause equilibrium models cannot describe steepconcentration gradients). In a chemically induced–proximity event, one member of a ternary com-plex, [A], is freely diffusing, while the other, [B],is localized (at the cell membrane, on chromatin,etc.). The addition of a chemical dimerizer createsa concentration gradient of the complete complex,

[AB*], with a maximum concentration at the re-cruitment site (Fig. 2A).The reaction-diffusion equation is as follows:

@uðx; tÞ@t

¼ D@2uðx; tÞ

@x2þ kuðx; tÞ

From Fick’s laws of diffusion, the flux of thesubstance at position (x) is proportional to theconcentration gradient, and the change in con-centration with respect to time (@u/@t) is related

through the differential equation @2udx2. The rate of

concentration changes (@u/@t) is also impacted bythe reaction rate, ku(x, t). In the absence of a chem-ical dimerizer, dimerizing proteins [A] and [B]have little-to-no binding affinity for each other(Fig. 2B). Although they freely collide, their rate ofdiffusion dominates over the binding rate. By con-trast, in the presence of a CIP (Fig. 2B), reaction isfaster than diffusion. As a result, the concentrationof bound [AB*] near the recruitment site is fargreater than the freely diffusing condition, whichcreates a virtual cloud of molecules to amplify theeffects of proximitywhile relieving steric constraints.Although it is tempting to solely credit the

superiority of chemical dimerization to purekinetics, the thermodynamic contributions ofthe system should not be understated (Fig. 2C).By increasing the effective molarity of a sub-strate, the cell is relieving the cost of translational(x, y, z) entropy (Fig. 2C). The use of a CIP mini-

mizes the relative configurational entropy of thesystem by reducing the possible collision anglesrelative to freely diffusing molecules. As a result,the CIPprovides both kinetic and thermodynamicadvantages by increasing the probability of inter-actions through effective concentration and byminimizing translational or rotational entropy.Despite the early characterization of binary-

complex equilibria in 1916 by Irving Langmuir(19), a mathematical description of a three-bodysystem, for example, FKBP-rapamycin-FRB, inequilibria (Fig. 2D) has only recently been de-scribed (20). Spiegel and others developed amathematical framework for ternary-complexformation that used measurable parameters(analogous to total concentration and dissoci-ation constants) to define the maximum concen-tration of dimerized complexes, [AB*]max (20).Their framework was extended to several biologicsystems and cooperative ternary complexes,including the TCR. For systems such as FKBP-rapamycin-FRB, where the dissociation constantKd values have been rigorously defined (21), morecomplete descriptions of the kinetics of ternarysystems may prove useful when characterizingthe fundamental processes governed by proximity.

Using induced proximity to explorebiologic mechanisms: Biologic mimicry

Arguably the major contribution of chemicallyinduced proximity is the ability to rapidly initiate

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 2 of 9

Fig. 1. The evolution of systems for CIPs. Protein targets andchemical ligands are shown for CIP systems. Proteins are representedas ribbon diagrams from available crystal structures, with endogenousmonomeric functions indicated. Chemical ligands are represented inbound conformations docked with protein targets and also represented

separately as individual structures. Gray shading of the chemicalstructures is divided to annotate specific structural moieties associatedwith molecular recognition of annotated protein targets. CIP systemsare represented from left to right in approximate order of development.Me, methyl group; R, linker moiety.

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

and track theminute-by-minute consequences ofa biochemical process in living cells. This allowsprecise kinetic studies, construction of syntheticregulatory circuits, and analysis of cellular mem-ory and places stringent limitations on causalitywithin genetic and biochemical networks.

Induced proximity in signal transductionand transcription

Chemically induced proximity fueled conceptualadvances in understanding signaling, includingthe role of proximity, the ordering of biochemicalevents, and the intersection with transcription.By using molecules that did not show allostericeffects upon binding their ligands by crystallog-raphy, it was found that receptor signaling couldbe induced by chemically mediated proximity(Fig. 3A), initially at the TCR (3) and later for ahost of different receptors. Dimerizing TCR-signaling components at the membrane withFK1012 revealed that dimerization was suffi-cient to initiate downstreamTCRsignaling events

(22, 23). In addition, recruiting the guanine nu-cleotide exchange factor Sos to the membranerevealed that Sos proximity could induce Rassignaling and that a major role of the linkingmolecule Grb-2 was to increase effective localconcentration (24).Temporal analysis of the biochemical conse-

quences of proximity defined the order of com-plex signalingmechanisms using both linear andparallel steps in a pathway. Membrane-inducedproximity of zeta chain–associated protein kinase(ZAP70) helped place its function in the TCRsignaling cascade, as did similar approaches forsignal components downstream of death andgrowth factor receptors (25). Activation of indi-vidual signaling molecules, not possible with lig-ands that induce several pathways, unveiled the“AND gate” function of several transcription fac-tors, including NFAT, meaning that two signalinginputs are required for a robust transcriptionaloutput. For example, isolated Ras activation couldnot activate NFAT nor could isolated Ca2+ signal-

ing. SimultaneousRas andCa2+ signalswere essen-tial forNFAT-dependent transcriptionandprovideda check on inappropriate gene activation (24).The role of proximity in transcriptional regu-

lation became clear with the early understandingof the spatial organization of promoters and theproteins bound by them. However, chemically in-duced dimerization allowed the examination ofthe in vivo kinetics in yeast, flies, and mammals.Chemical recruitment of transcriptional activatorsled to the finding that transcriptional activationcan be accomplished through proximity on timescales of minutes, rather than hours or days(Fig. 3B) (6), helping investigators to temporallyorder events in the complex sequence leading totranscriptional activation.An important feature of a CIP is its rapid re-

versibility (by small-molecule washout with com-petitive inhibitors), which enables the study ofmolecular memory in cells. To enhance revers-ibility, nontoxic FK506 analogs were developedto competitively wash out synthetic dimerizers.The first competitive inhibitor of dimerization(FK506M) (3) was used to demonstrate thatdimerizer-mediated transcription was rapidly re-versible and induced no stable memory (Fig. 3B).However, developing a system for carrying outorder-of-addition and co-occupancy of activatorsand repressors lead to the discovery that tran-scription could persist in certain contexts in yeasteven after the activator was released (26). Thisindicated that memory was hardwired intothese systems, by virtue of repressor resistance.More-refined analysis of transcriptional memoryemerged from later studies of epigenetic regu-lators (see section “Chromatin regulation” belowand Fig. 3E).Induced proximity has also been pivotal for

understanding the kinetics of transcriptional reg-ulation in individual cells. In both yeast andhuman cells, transcription at a single allele wasinduced in an all-or-none quantalmanner (6, 27).

CRISPR and CIP-regulated transcription

Recent advances with CRISPR-Cas9 (28) haveushered in a new era of CIP transcriptional reg-ulation (29, 30). Zhang and others developed arapamycin-inducible assembly of enzymaticallydead Cas9 (dCas9) and locus-specific guide RNAs(Fig. 3B) (31). Recently, other dCas9 fusions(Fig. 3B) were found to be highly compatiblewith a variety of CIP systems (32, 33).The dCas9-based dimerizer systems allow com-

binatorial recruitment as well as ordered recruit-ment of activators and repressors, which haveenabled studies of synergy and antagonism. Forexample, proximity-induced formation of repres-sive transcriptional states by recruiting a KRABrepression domain had a deterministic silencingeffect on transcription, even with co-recruitmentof an activator (32).

Protein folding and localization

Regulated compartmentalization of molecules isa common biological process easily mimickedby CIPs. Schreiber and others used a syntheticheterodimer, FKCsA (Fig. 1) (34), which targeted

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 3 of 9

Fig. 2. Modeling reaction kinetics associated with systems of induced proximity. (A) Thedifferential concentration with respect to time is explained by the changes in the rate of diffusion andthe binding kinetics of the dimerizer system. (B) Changes in concentrations of monomeric anddimeric complexes are dependent on rate of reaction and rate of diffusion, defined by the distance tothe site of recruitment. With no chemical induction or high Kd, formation of ternary complexes isdetermined by the rate of diffusion, as ku(x, t) approaches zero. With chemical induction and lowKd, induced ternary-complex formation is strongly dependent on the rate of the reaction ku(x,t),which dominates the rate of diffusion. Direct-fusion systems are exclusively localized at the site ofrecruitment as the reaction rate approaches infinity and dominates the rate of diffusion. Inprotein-dimerizer interactions, these complexes are designated by *. (C) Thermodynamic contributionsto chemically induced dimerization include minimizing translational and rotational entropy. Multistatebinding equilibria associated with initial binding of a bifunctional dimerizer molecule (hexagons) torespective targets by forming unstable binary complexes that form composite surfaces and rapidlyassemble ternary complexes. Arrows show direction of movement or rotation. (D) Kinetics ofternary-complex assembly can be described by three-body binding equilibria.

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

FKBP and prolyl isomerase CyP without bindingcalcineurin, to rapidly induce nuclear translocationof CyP-tagged green fluorescent protein (GFP)with a nuclear-localized NLS-FKBP. A CIP wasalso used to rapidly export proteins from thenucleus, thereby reversibly inactivating them(35). Furthermore, this approach proved highlyeffective in yeast for inactivating nuclear pro-teins by shuttling them out of the nucleus usingthe “anchor-away” system (36–39). Anchor awayis now frequently used to rapidly inactivate andreactivate nuclear proteins to understand theirdirect actions.

By expanding CIP localization beyond thenucleus to other organelles, Rivera and colleaguessought to activate specific secretory pathways inthe endoplasmic reticulum (ER) for therapeuticpurposes, building upon two critical advances(40). First, a multimer-forming, conditional aggre-gator of FKBP12(F36M) was found to be retainedin the ER (Fig. 3C) in the absence of chemicalligands (AP22542, APAP21998). Second, a Golgi-specific protease (furin) was harnessed to targeta cleavage site (FCS) engineered into fusion pro-teins with human growth hormone or insulin.By coupling these two advances, it was demon-

strated that a CIP could both simultaneouslycleave FKBP and, by resolving the aggregation,induce protein secretion (40). Furthermore, theseligands induced insulin secretion in hypergly-cemic, FKBP(F36M)-FCS-insulin transgenic mice.This study showcased the clinical potential ofCIPs for gene therapy applications, in addition toproviding new insights into the secretory pathway.Chemical dimerizers were further utilized to

investigate how Golgi membranes associate withthe ER during cell division (41). To investigatesecretory mechanisms of Golgi-ER interactionduring the cell cycle, the authors expressed fu-sions of FKBP-GFP with sialyltransferase (ST;Golgi specific), and FRAP-HA with the humaninvariant-chain protein (li; ER specific) (41).When coexpressed in COS-7 cells, the proteinsremained associated within their respective cel-lular compartments as monomers, even uponthe addition of chemical dimerizer. Notably, whentreated with brefeldin A, a small molecule thatinduces Golgi-ER membrane fusion, the authorsobserved rapamycin-dependent colocalization ofST and li. This unexpected finding, that the Golgiand ER exhibit spatial independence during celldivision, demonstrates how CIPs continue to re-veal previously unknown aspects of seeminglywell-characterized biologic mechanisms.Svoboda and others developed an ingenious

CIP approach to tether synaptic vesicle proteins.This allowed for inducible and reversible activa-tion of synaptic transmission in neurons (Fig. 3C)(42). Inmotor neurons expressing vesicle-associatedmembrane protein (VAMP) or synaptobrevin-FKBP(F36V), the authors inhibited 50 to 100% ofsynaptic transmission in minutes using the syn-thetic dimerizer AP20187 (Fig. 1). This systemwasextended to both cultured neurons ex vivo andmotor neurons in vivo. Furthermore, in Purkinjeneurons of living mice, dimerization of VAMP–Syb-FKBP(F36V) with AP20187 induced function-al ataxia during learned balancing tasks (42). Thisstudy highlights the potential for CIPs in under-standing neuron function and complements in-vasive optogenetic systems.The mechanism of Ca2+ entry was probed by

Lewis and others through stromal interactionmolecule 1 (STIM1) oligomerization (43). Theauthors demonstrated that the rapalogue AP21967could rapidly oligomerize STIM1-FKBP and FRB-STIM1 and localize the complex to the cell pe-riphery. This activatedCRAC(Ca2+ release–activatedCa2+) channel currents and revealed that inducedoligomerization of STIM1 was sufficient for cal-cium entry via CRAC channels.The expanded chemical repertoire (includ-

ing new orthogonal rapalogs that were specifi-cally tuned to FRBmutants) heavily influencedthe multiplex ability of subcellular-localizationstudies (44). First, a triple-mutant FRB [residueLys2095→Pro2095 (Lys2095Pro), Thr2098Leu,Trp2101Phe], denoted FRB*, was developed toselectively form a ternary complex with FKBPand C20-methallylrapamycin (C20-MaRap) (45).Through utilization of a CIP transcriptional-reporter screen, it was found that by introducingrapalogs and respective FRB mutants, precise

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 4 of 9

Fig. 3. Chemical induction of proximity is sufficient for the regulation of diverse cellularprocesses. Induced proximity has been shown to regulate initiation of transcription, signaling cascades,chromatin dynamics, proteasomal degradation, and subcellular localization. (A) Induced proximityhas been systematically explored to bypass Tcell antigen receptor activation and for synthetic inductionof a variety of signaling cascades. (B) CIPs have been developed for rapid induction of transcriptionalactivation (VP16) and repression (KRAB or HP1) using DNA binding domains (DBDs) as well as CIP ofsplit–CRISPR-Cas proteins or CIP recruitment of activators or repressors through CRISPR-Cas9systems. (C) CIP has been used for rapid protein localization, including nuclear import and export,localization to components of the secretory pathway, synaptic vesicles, and mitochondria. (D) Rapidproximity-based protein degradation is achieved through bifunctional molecule–mediated recruitmentof E3 ubiquitin ligase complexes (complex composed of E2, ROC1, CUL4A, DDB1, and CRBN).With arelated approach, auxin can induce ubiquitin-mediated degradation through recruitment of the TIR1-Cul1complex. (E) Induced proximity has been used for rapid induction of activated chromatin states throughrecruitment of ATP-dependent remodeling complexes and induction of repressive chromatin statesthrough HP1-mediated heterochromatin formation.

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

measurements could be made to define new pat-terns of specificity. By using these new orthogonaldimerizers, simultaneous expression of nuclearexporting FRB*(LT) and nuclear localizing FRB*(LW) provided a platform for push-pull control ofFKBP–GSK3b (glycogen synthasekinase 3b),whichwas modulated by treatment with the appropri-ate orthogonal rapalog (44).Modulating protein structure is another critical

component of posttranslational modification thathas been explored through proximity-based ap-proaches. Muir and others developed a facile ap-proach to conditional protein structural variation.Using rapamycin, they were able tomimic naturalprotein splicing with proximity-based inteincleavage (46).Recently, Ballister and colleagues developed a

clever light-induced proximity system—defininga photon as the smallest dimerizer (47). Modify-ing a previous bifunctional bis-methotrexatedimerization system (48), the authors labeled thedihydrofolate reductase (DHFR) ligand with aphotocleavable moiety, which blocked the requi-site DHFR-interacting surfaces in the absence ofirradiation. Tagging the photocleavable moietywith a HaloTag linker formed an irreversibleadduct with a Haloenzyme allowing for selective,light-inducible Halo-tagging and subcellular re-localization of DHFR upon irradiation. Further-more, photoinduced DHFR relocalization wasextended to the centromere, kinetochore, cen-trosome, andmitochondriawithCENP-Halo,Nuf2-Halo, AKAP9-Halo, and ActA-Halo, respectively.

Protein degradation

Loss-of-function studies have been the mainstayof genetics but are plagued by the slow loss of

protein, allowing the accumulation of compen-satory and indirect responses clouding mechanis-tic interpretation. CIP-regulated protein stabilitywas developed to circumvent these classic prob-lems. By using C20-MaRap, it was determinedthat the stability of FRB* fusions (44, 45) was de-pendent on formation of the FRB*–C20-MaRap–FKBP ternary complex. To investigate the functionof GSK3b in developing mice, FRB*was knockedinto the endogenousGSK3b gene. Because expres-sed GSK3b could only be stabilized in the pres-ence of C20-MaRap andwas otherwise degraded,dosing for short periods allowed the authors todefine separate, discrete periods of developmentduring which the gene executed its function inskeletogenesis and palate development (49).Wandless and others found that double-mutant

FKBP(Phe36Val, Leu106Pro) could also be used asa conditionally stabilizing allele, which allowedfor rapid in vitro degradation of target proteins(21). The synthetic dimerizer (Shield-1) used inthese studies was degraded in minutes in cell cul-ture, allowing rapid reversal of the reaction.To regulate proteasome-mediated degradation,

CIP systems were developed to target chimeric E3ligase complexes. The TIR1 receptor–auxin (Fig. 1)degradation pathway inArabidopsis, which utilizesthe dimerizer indole-3-acetic acid (IAA; auxin), in-duces dimerization of the TIR1-SCF E3 ubiquitinligase complex with an auxin-inducible degron(AID) (50, 51). In nonplant systems, TIR1 was suc-cessfully reconstituted into endogenous E3 ligasecomplexes to selectively recruit AID-fusion pro-teins to the Cul1 complex (50). This resulted inauxin-mediated ubiquitination of the AID fusionand rapid proteasomal degradation (Fig. 3D). Later,a short 44–amino acid tag referred to as IAA17

(AID*) was developed to expand the utility of theauxin system (52). The auxin-degron (AID*) sys-tem has been used to regulate kinases and es-sential genes that lack selective inhibitors. Theessential Plk4 kinase, which is associated withtumor suppression (53, 54), was degraded withAID-fusion transgenes (55) and homozygousknockins (56) to reveal that Plk4 positively reg-ulates centriole duplication in a reversible anddosage-dependent manner. The auxin system hasenabled rapid degradation of a wide variety oftargets across many species (55, 57–59). Recently,auxin-mediated degradation of the transcriptionfactor CTCF (60) has differentiated its roles in localtopologically associating domain (TAD) structurefrom chromosomal-compartment architecture.

Chromatin regulation

The immense complexity of chromatin, with itsmany developmentally specific histonemodifica-tions, topology, long-range interactions, variegatedDNAmethylation, anduncharacterized chromatincomponents, has proven to be a formidable targetfor investigation. The limitations encounteredin formation of chromatin in vitro have becomeapparent (61). To circumvent these challenges, aCIP technique (CiA, chromatin in vivo assay) wasdeveloped to study chromatin in all its topolog-ical, biochemical, and developmental diversity(62). With CiA, one can “chemically pipette” achromatin regulator of interest into essentiallyany locus in the genome of any cell type (Fig. 3E).TheCiA systemwas first used to study chromatin-

based memory. Recruiting heterochromatin pro-tein 1a (HP1a) to the active Oct4 (transcriptionfactor) locus inmouse embryonic stemcells (ESCs),resulted in an expanding domain of repression

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 5 of 9

Fig. 4. Ubiquitin ligasecomplexes rapidly degradeoncogenic protein targets.(A) The CBRN-CUL4A ubiq-uitin ligase complex can berecruited to BCR-ABL andBRD4 with thalidomideconjugated to bosutinib andJQ1, respectively. Inducedproximity rapidly degradesthese targets, which areknown to drive chronic mye-logenous leukemia (BCR-ABL) and acute myeloid leu-kemia (BRD4). (B) The VHLligand fused to desatinibor JQ1 can efficiently recruitthe VHL-Cullin2 ubiquitinligase complex to ABL andBRD4, respectively. In eachcase, rapid degradation ofthe oncogenic targets resultsin inhibition of cancer growth.

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

that silenced Oct4. By washing out the CIP, itwas observed that H3K9me3 (trimethylatedhistone H3 on lysine 9) islands are stable fordays after rapamycin washout but that theeffect could be rapidly reversed by initiatingtranscription by plant hormone abscisic acid(ABA)–mediated recruitment of the transcriptionalfactor VP-16 (62). Furthermore, the ability torapidly controlH3K9me3-based heterochromatinpermitted mathematical modeling, which putforth a “balanced intrinsic reaction rate”mod-el for the propagation of H3K9me3 repressionon the basis of kinetic parameters (62, 63). Inthis model, rates of addition and removal de-termine the propagation rate of H3K9me3 andaccurately predict 99% of the H3K9 domainsover the murine ESC genome.One of the most persistent problems in epi-

genetics has been understanding the placementand stability of polycomb repressive complexes(PRCs). In 1988, mutations in the Brahma gene,which encodes an adenosine triphosphate (ATP)–dependent chromatin remodeler, were found torepress mutations in the PRC1 complex, indicat-ing that these two chromatin regulators opposedone another (64). The mechanismwas elusive be-cause of the inability to form PRC-repressed het-erochromatin in vitro. Tounderstand this problem,the SWI/SNF (switch-sucrose nonfermentable)

or BAF (Brahma/Brg associated factor) complexwas recruited to a polycomb-repressed promoterwith the CiA system (65, 66). PRC1 eviction oc-curred in minutes, followed by PRC2 eviction.The rapid action of BAF complexes led to thefinding that they directly bind and release PRC1by an ATP-dependent mechanism (65, 66). Fur-thermore, heterozygous expression of cancer muta-tionsofBrg, the regulatory adenosine triphosphatase(ATPase) of BAF, lead to polycomb accumulation,extending the CiA results to the genome. NewCIP-dCas9 systems for manipulating chromatinarchitecture (33, 67) will likely prove critical inuncovering chromatin regulatorymechanisms ina host of genomic contexts.CIPs provided additional insights into the

dissolution and formation of heterochromatinby the observation that recruitment of the BAFcomplex lead to accumulation of TopoIIa bind-ing at the recruitment site, as suggested from anearlier study (68). Unexpectedly, TopoIIa functionwas found to be essential for both the dissolutionof heterochromatin after recruiting BAF and theformation of heterochromatin after releasingBAF (69). Remarkably, the strand-cleaved reactionintermediate was found at the precise time andposition of heterochromatin formation and dis-solution. These studies, using rapid reversible co-recruitment of TopoIIa and BAF, indicated that

decatenation is essential for the regulation ofheterochromatin.

Chromosomal dynamics

To assess the mechanisms of DNA associationwith the cohesin complex during cell division,Nasmyth and others used dimerizers to “lock”the Smc1-Smc3 complex in place during specificwindows of the cell cycle in yeast (70). By releasingyeast from G1 (prereplicative phase) arrest, withor without the conditional dimerization of Smc1and Smc3, it was discovered that the Smc1-Smc3complexmust open duringmitosis and that thiswas necessary for chromatid cohesion.A similar CIP-based approach was used to

understand the role of Scc1 in sister chromatinassociation. Relocalization and inhibition of Scc1-FRB by rapamycin with ribosomal protein anchorRPL13A-FKBP demonstrated that 30% of sisterchromatid association was disrupted by anchor-ing away Scc1 in yeast (36).

Induced proximity in medicineDegrading or inactivating pathogenicproteins

The above studies, directed primarily at dissect-ing biologic mechanisms, demonstrate howwidelythis methodology can be used. However, the ap-plication to the treatment of disease had beenhampered by the requirement that proteinsmustbe tagged with dimerizing peptides. Graef andothers reasoned that induced proximity of en-dogenous, unmodified proteins would be wide-ly useful. To make a potential therapeutic forAlzheimer’s disease, they synthesized a moleculethat bound both FKBP (SLF, synthetic ligand ofFKBP) and the pathogenic b-amyloid (Ab) peptide(CR, Congo red) (71). Their two-sided molecule(SLF-CR) showed activity in in vitro assays ofAb aggregation but was too toxic to be used as atherapeutic. A similar approach also extended thehalf-life of an HIV protease inhibitor by causing itto remain intracellular and protected (72).Although neither of these bifunctional mole-

cules had good pharmacologic characteristics,this conceptual advance precipitated a wave ofefforts to extend proximity-inducing moleculesto many other medical problems. One of the firstof these was designed to stabilize the pathogenicaggregation of proteins. Transthyretin (TTR) canproduce aggregating amyloid fibrils and causesamyloidoses, including cardiomyopathies suchas senile system amyloidoses, familial amyloidcardiomyopathy, and familial amyloid poly-neuropathy. The development of bifunctionalstabilizers of TTR, such as AG10 by Graef andothers (73), provides a promising candidate in pre-venting the progression of diseases associatedwith amyloid aggregation.In many diseases, pathogenic proteins arise

from mutation, recombination, or stable alloste-ric modification. What if these culprits could bedegraded by induced proximity? In 2010, CRBN,a component of the DDBI-CRBN E3 ubiquitinligase complex (Fig. 3D), was found to be theprimary molecular target of thalidomide andrelated molecules (IMiDs, immunomodulatory

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 6 of 9

Fig. 5. CAR T cell therapeutic applications of CIPs. (A) Engineered safety switches usingATTAC systems with AP1903. (B) An inactive engineered CAR T cell receptor and (C) an activeengineered CAR T cell receptor binding its cognate antigen. (D) AP1903-induced caspasedimerization and activation allows for rapid apoptosis of CAR T cells to prevent complicationsthat may arise from transplant.

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

drugs) (74). Several years later, elegant struc-tural studies for IKAROS family transcriptionfactors (IKZF1, IKZF3) indicated that ubiquitinbinding and targeting was IMiD dependent (75).These insights, coupled with the knowledge thatthalidomide differed from analogs lenalidomideand pomalidomide through a single C-4 anilinesubstitution (75), provided a clear path to con-jugate new ligands for the purposes of proximity-based protein degradation.To investigate the potential applications of

CRBN-IMiDs, the Bradner and Crews labs de-veloped strategies to conjugate BRD4-targetingcell-permeable small molecule JQ1 (76) with thalid-omide and investigated its potential as an induc-ible proximity-based ubiquitinase (Fig. 4) (77, 78).These studies were precipitated by the observa-tion that IMiDs could bind directly to CRBNwithout inhibiting the associated ubiquitin ligasecomplex. In the first study, the phthalimide-JQ1conjugate (dBET1) induced ubiquitin-mediateddegradation of BRD4 on a time scale of hours,with a mechanism analogous to auxin-based deg-radation systems (without requiring genetic ma-nipulations). The same C-4 phthalimide linkagewas used to construct thalidomide-SLF conjugates(d-FKBP-1) for rapid and selective ubiquitin-mediated degradation of FKBP12. This demon-strated that these bifunctional conjugates werehighly selective and had activity in human cells.By using a longer polyethylene glycol (PEG)–based linker attached to phthalimide, Crews andothers simultaneously found that BRD4 could bedegraded using both CRBN and von Hippel–Lindau tumor suppressor (VHL) ubiquitin ligase–targeting ligands with their strategy termedproteolysis targeting chimeras (PROTACs) (Fig. 4).Their bifunctional PROTACswere capable of rap-idly recruiting the VHL-associated ubiquitin li-gase complexes by using ligands specific to bothestrogen-related receptor alpha (ERRa) and theserine-threonine protein kinase RIPK2 with no-tably high selectivity and activity in livemice (79).In an important follow-up study, Bradner and

others identifiedENL, bearing aYEATSacetylated-lysine reader domain, as the product of an essen-tial gene in a human acute myeloid leukemia(AML) model system with a mixed-lineage leu-kemia (MLL)–AF4 translocation (MV4;11) (80).To understand the function of ENL as a poten-tial driver in AML, the authors expressed theENL-FKBP12(F36V) protein in an ENL-deficient(ENL−/−) MV4;11 cell line and used SLF (with nocalcineurin or mTOR inhibition) conjugated tophthalimide (dTAG-13). Selective degradation ofENL with dTAG-13 resulted in decreased expres-sion of AML drivers, including MYC, HOXA10,andMYB, and substantial reduction in elongationfactors AFF9 and CDK9. This revealed that ENLmay drive leukemogenesis through binding andelongation of canonical AML targets and dem-onstrated a creative use of degrading-CIPs withtherapeutic potential.Inhibition of BRD4, which results in repression

of c-Myc activity, has been proposed as a thera-peutic strategy in a host of diseases, includingAML, acute lymphoblastic leukemia, NUTmidline

carcinoma, and HIV (76, 81, 82). Concomitantly,Crews and others, aware of the limitations ofpeptide-based degradation strategies, indepen-dently identified potent, small molecules target-ing VHL E3 ubiquitin ligase (83). By utilizing thisVHL-targeting strategy, dimeric ligands haveselectively degraded ERRa RIPK2, which is in-volved in nuclear factor kB (NF-kB) andmitogen-activated protein kinase (MAPK) activation, andBRD4 (79, 84, 85). Furthermore, proximity-induceddegradation provides a platform for rapid iterationthrough combination. Crews and others recentlydemonstrated that by varying the E3 ligase targetand the functional protein-targeting warhead, theselectivity of known tyrosine kinase inhibitors (TKIs)could be improved (85). By modulating knownTKIs—imatinib, dasatinib, or bosutinib—and theE3 ligase target, these small molecules degradedboth c-ABL and the oncogenic BCR-ABL (Fig. 4),with varying degrees of specificity (85). Even thoughTKIs have proven to be immensely successful inthe treatment of chronic myelogenous leukemia,it remains a lifelong condition. One hypothesisfor persistent leukemic cells is that the patho-genesis is not entirely dependent on kinase ac-tivity and that BCR-ABL may play a scaffoldingrole in signaling (86). As such, proximity-induceddegradation may not only provide a mechanismfor inhibition of oncogenic proteins [with efficacycomparable toRNA interference (RNAi) or CRISPR,but without the immunogenicity of Cas9] butmay also provide cures for diseases for which pre-sent catalytic inhibitors are inadequate. This strat-egymay revive the imperfect chemical probes thatbound the protein of interest but failed to delivercures for critical therapeutic targets.

Induced proximity in cellular therapies

Gene therapies require the delivery of preciseamounts of therapeutic proteins at specific timesas well as a humanized system to prevent im-mune rejection of the engineered cells. On thebasis of earlier FK1012-mediated transcriptionalactivation studies (6), Clackson and others devel-oped a completely humanized delivery systemthat provides long-term, regulated expressionin primates (87–89). A major challenge for anymethod of regulated gene expression is the steepdose-response curve induced by rapamycin. Theuse of nontoxic dimerizers such as abscisic acidprovide a more graded dose response (13) andcould be useful for precise dosage control.Certain therapeutic strategies require removal

of pathogenic cell types. Early studies demon-strated that dimerizing the intracellular domainof the Fas receptor or other death-signaling mol-ecules could accomplish this goal (90–93). By usinga technique called apoptosis through targeted ac-tivation of caspase 8 (ATTAC) (Fig. 5A), an animalmodel was developed to study obesity and glucose-stimulated insulin secretion. This CIP “suicide-switch” strategy was extended to a senescentcell–clearing mouse to study age-related path-ologies (94). Baker and colleagues observed thatCIP-mediated clearance of senescent cells ex-tended health and life span of normal tissues.Furthermore, with precise temporal control, the

ATTAC system attenuated the progression of age-related diseases (94, 95).Cellular therapies remain one of the most

hopeful strategies; however, concerns remainaround potential off-target damage or possiblemalignancy that may result from genomic inte-gration of an introduced gene. Although allogenictransplantation of hematopoietic stem cells is aneffective leukemia treatment, positive benefitsof this therapy are often counteracted by graft-versus-host disease (GVHD). To circumvent this,a CIP “safety switch”was developed to selectivelyinduce apoptosis in hematopoietic transplants inthe case that severe GVHD arises in patients (96).In a small clinical trial, patients withGVHDweretreated with AP1903 (Fig. 1), a bioinert analog ofFK1012 (97). AP1903 selectively eliminated 90% ofthemodified T cells within 30min and eliminatedGVHDwithout recurrence (96). Subsequent studiesfurther demonstrated the usefulness of this safetyswitch in long-termGVHDcomplications (98, 99).CIP safety switches may also prove to be a

promising strategy for mitigating side effects ofcancer immunotherapy treatments. In a recentstudy in humanized mice, T cells were simulta-neously modified with a chimeric antigen re-ceptor (CAR) and the iCaspase9 safety switch(Fig. 5, B to D). Despite the effectiveness of CART cell therapies for treating B cell malignancies(including acute lymphoblastic leukemia andlymphomas) (100, 101), possible side effects canbe severe. Treating mice with a CD19–FKBP-iCaspase9 T cell therapy provided two advan-tages. First, induced dimerization of the caspase9 protein provides a built-in temporally controlledmechanism to ablate harsh side effects in patients,such as cytokine release syndrome or B cell apla-sia. Second, selective apoptosis provides a mech-anism to eliminate transplanted T cells in acontrolled manner that allows for patient-specificresponses in the clinic, along with the ability torepopulate a patient’s own immunity (102).

Summary

The application of chemically induced proximityto elucidatebiologicmechanismscontinues togrowwith recent advances in understanding epigeneticregulation, chromosomal dynamics, and topology.However, the use of this mechanism in treatmentof disease is still in its embryonic form. Bifunc-tionalmolecules that use induced proximity for theelimination of pathogenic proteins and aggregatedproteins and to control subcellular localizationare likely to make a substantial impact on thetreatment of disease in the near future. The de-velopment of totally humanized systems for geneand cellular therapy is now under clinical inves-tigation and showing promise. Small moleculesthat capture the universal biologic regulatorymechanism of induced proximity will likely havemany other unanticipated uses and provide aplayground for our imaginations.

REFERENCES AND NOTES

1. Z. Songyang et al., SH2 domains recognize specificphosphopeptide sequences. Cell 72, 767–778 (1993).doi: 10.1103/RevModPhys.62.251; pmid: 7680959

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 7 of 9

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

2. P. Hänggi, P. Talkner, M. Borkovec, Reaction-rate theory: Fiftyyears after Kramers. Rev. Mod. Phys. 62, 251–341 (1990).doi: 10.1103/RevModPhys.62.251

3. D. M. Spencer, T. J. Wandless, S. L. Schreiber, G. R. Crabtree,Controlling signal transduction with synthetic ligands.Science 262, 1019–1024 (1993). doi: 10.1126/science.7694365; pmid: 7694365

4. Z. Luo et al., Oligomerization activates c-Raf-1 through a Ras-dependent mechanism. Nature 383, 181–185 (1996).doi: 10.1038/383181a0; pmid: 8774885

5. P. J. Belshaw, D. M. Spencer, G. R. Crabtree, S. L. Schreiber,Controlling programmed cell death with a cyclophilin-cyclosporin-based chemical inducer of dimerization. Chem. Biol.3, 731–738 (1996). doi: 10.1016/S1074-5521(96)90249-5;pmid: 8939689

6. S. N. Ho, S. R. Biggar, D. M. Spencer, S. L. Schreiber,G. R. Crabtree, Dimeric ligands define a role fortranscriptional activation domains in reinitiation. Nature 382,822–826 (1996). doi: 10.1038/382822a0; pmid: 8752278

7. M. J. Holliday, C. Camilloni, G. S. Armstrong, M. Vendruscolo,E. Z. Eisenmesser, Networks of dynamic allostery regulateenzyme function. Structure 25, 276–286 (2017).doi: 10.1016/j.str.2016.12.003; pmid: 28089447

8. J. Kallen et al., Structure of human cyclophilin and its bindingsite for cyclosporin A determined by x-ray crystallographyand NMR spectroscopy. Nature 353, 276–279 (1991).doi: 10.1038/353276a0; pmid: 1896075

9. N. A. Clipstone, G. R. Crabtree, Identification of calcineurin asa key signalling enzyme in T-lymphocyte activation. Nature 357,695–697 (1992). doi: 10.1038/357695a0; pmid: 1377362

10. X. F. Zheng, D. Florentino, J. Chen, G. R. Crabtree,S. L. Schreiber, TOR kinase domains are required for twodistinct functions, only one of which is inhibited byrapamycin. Cell 82, 121–130 (1995). doi: 10.1016/0092-8674(95)90058-6; pmid: 7606777

11. S. W. Michnick, M. K. Rosen, T. J. Wandless, M. Karplus,S. L. Schreiber, Solution structure of FKBP, a rotamaseenzyme and receptor for FK506 and rapamycin. Science 252,836–839 (1991). doi: 10.1126/science.1709301;pmid: 1709301

12. S. R. Cutler, P. L. Rodriguez, R. R. Finkelstein, S. R. Abrams,Abscisic acid: Emergence of a core signaling network. Annu.Rev. Plant Biol. 61, 651–679 (2010). doi: 10.1146/annurev-arplant-042809-112122; pmid: 20192755

13. F. S. Liang, W. Q. Ho, G. R. Crabtree, Engineering the ABAplant stress pathway for regulation of induced proximity. Sci.Signal. 4, rs2 (2011). doi: 10.1126/scisignal.2001449;pmid: 21406691

14. T. Miyamoto et al., Rapid and orthogonal logic gating with agibberellin-induced dimerization system. Nat. Chem. Biol. 8,465–470 (2012). doi: 10.1038/nchembio.922;pmid: 22446836

15. S. An, R. Kumar, E. D. Sheets, S. J. Benkovic, Reversiblecompartmentalization of de novo purine biosyntheticcomplexes in living cells. Science 320, 103–106 (2008).doi: 10.1126/science.1152241; pmid: 18388293

16. S. Y. Tsuji, D. E. Cane, C. Khosla, Selective protein-proteininteractions direct channeling of intermediates betweenpolyketide synthase modules. Biochemistry 40, 2326–2331(2001). doi: 10.1021/bi002463n; pmid: 11327852

17. R. J. Conrado, T. J. Mansell, J. D. Varner, M. P. DeLisa,Stochastic reaction-diffusion simulation of enzymecompartmentalization reveals improved catalytic efficiencyfor a synthetic metabolic pathway. Metab. Eng. 9, 355–363(2007). doi: 10.1016/j.ymben.2007.05.002; pmid: 17601761

18. S. B. van Albada, P. R. ten Wolde, Enzyme localization candrastically affect signal amplification in signal transductionpathways. PLOS Comput. Biol. 3, e195 (2007).doi: 10.1371/journal.pcbi.0030195; pmid: 17937496

19. I. Langmuir, The constitution and fundamental properties ofsolids and liquids. Part I. Solids. J. Am. Chem. Soc. 38,2221–2295 (1916). doi: 10.1021/ja02268a002

20. E. F. Douglass Jr., C. J. Miller, G. Sparer, H. Shapiro,D. A. Spiegel, A comprehensive mathematical model forthree-body binding equilibria. J. Am. Chem. Soc. 135,6092–6099 (2013). doi: 10.1021/ja311795d; pmid: 23544844

21. L. A. Banaszynski, L.-C. Chen, L. A. Maynard-Smith,A. G. L. Ooi, T. J. Wandless, A rapid, reversible, and tunablemethod to regulate protein function in living cells usingsynthetic small molecules. Cell 126, 995–1004 (2006).doi: 10.1016/j.cell.2006.07.025; pmid: 16959577

22. M. N. Pruschy et al., Mechanistic studies of a signalingpathway activated by the organic dimerizer FK1012. Chem.

Biol. 1, 163–172 (1994). doi: 10.1016/1074-5521(94)90006-X;pmid: 9383386

23. D. M. Spencer, I. Graef, D. J. Austin, S. L. Schreiber,G. R. Crabtree, A general strategy for producing conditionalalleles of Src-like tyrosine kinases. Proc. Natl. Acad. Sci.U.S.A. 92, 9805–9809 (1995). doi: 10.1073/pnas.92.21.9805;pmid: 7568222

24. L. J. Holsinger, D. M. Spencer, D. J. Austin, S. L. Schreiber,G. R. Crabtree, Signal transduction in T lymphocytes using aconditional allele of Sos. Proc. Natl. Acad. Sci. U.S.A. 92,9810–9814 (1995). doi: 10.1073/pnas.92.21.9810;pmid: 7568223

25. I. A. Graef, L. J. Holsinger, S. Diver, S. L. Schreiber,G. R. Crabtree, Proximity and orientation underlie signalingby the non-receptor tyrosine kinase ZAP70. EMBO J. 16,5618–5628 (1997). doi: 10.1093/emboj/16.18.5618;pmid: 9312021

26. S. R. Biggar, G. R. Crabtree, Chemically regulatedtranscription factors reveal the persistence of repressor-resistant transcription after disrupting activator function.J. Biol. Chem. 275, 25381–25390 (2000). doi: 10.1074/jbc.M002991200; pmid: 10801867

27. S. R. Biggar, G. R. Crabtree, Cell signaling can direct eitherbinary or graded transcriptional responses. EMBO J. 20,3167–3176 (2001). doi: 10.1093/emboj/20.12.3167;pmid: 11406593

28. J. A. Doudna, E. Charpentier, The new frontier of genomeengineering with CRISPR-Cas9. Science 346, 1258096(2014). doi: 10.1126/science.1258096; pmid: 25430774

29. D. M. Heery, E. Kalkhoven, S. Hoare, M. G. Parker, A signaturemotif in transcriptional co-activators mediates binding tonuclear receptors. Nature 387, 733–736 (1997).doi: 10.1038/42750; pmid: 9192902

30. J. D. Chen, R. M. Evans, A transcriptional co-repressor thatinteracts with nuclear hormone receptors. Nature 377,454–457 (1995). doi: 10.1038/377454a0; pmid: 7566127

31. B. Zetsche, S. E. Volz, F. Zhang, A split-Cas9 architecture forinducible genome editing and transcription modulation.Nat. Biotechnol. 33, 139–142 (2015). doi: 10.1038/nbt.3149;pmid: 25643054

32. Y. Gao et al., Complex transcriptional modulation withorthogonal and inducible dCas9 regulators. Nat. Methods 13,1043–1049 (2016). doi: 10.1038/nmeth.4042;pmid: 27776111

33. S. M. G. Braun et al., Rapid and reversible epigenome editingby endogenous chromatin regulators. Nat. Commun. 8, 560(2017). doi: 10.1038/s41467-017-00644-y; pmid: 28916764

34. P. J. Belshaw, S. N. Ho, G. R. Crabtree, S. L. Schreiber,Controlling protein association and subcellular localizationwith a synthetic ligand that induces heterodimerization ofproteins. Proc. Natl. Acad. Sci. U.S.A. 93, 4604–4607 (1996).doi: 10.1073/pnas.93.10.4604; pmid: 8643450

35. J. D. Klemm, C. R. Beals, G. R. Crabtree, Rapid targeting ofnuclear proteins to the cytoplasm. Curr. Biol. 7, 638–644(1997). doi: 10.1016/S0960-9822(06)00290-9;pmid: 9285717

36. H. Haruki, J. Nishikawa, U. K. Laemmli, The anchor-awaytechnique: Rapid, conditional establishment of yeast mutantphenotypes. Mol. Cell 31, 925–932 (2008). doi: 10.1016/j.molcel.2008.07.020; pmid: 18922474

37. X. Fan et al., Nucleosome depletion at yeast terminators isnot intrinsic and can occur by a transcriptional mechanismlinked to 3′-end formation. Proc. Natl. Acad. Sci. U.S.A. 107,17945–17950 (2010). doi: 10.1073/pnas.1012674107;pmid: 20921369

38. K. H. Wong, K. Struhl, The Cyc8-Tup1 complex inhibitstranscription primarily by masking the activation domain ofthe recruiting protein. Genes Dev. 25, 2525–2539 (2011).doi: 10.1101/gad.179275.111; pmid: 22156212

39. A. Charbin, C. Bouchoux, F. Uhlmann, Condensin aids sisterchromatid decatenation by topoisomerase II. Nucleic AcidsRes. 42, 340–348 (2014). doi: 10.1093/nar/gkt882;pmid: 24062159

40. V. M. Rivera et al., Regulation of protein secretion throughcontrolled aggregation in the endoplasmic reticulum. Science287, 826–830 (2000). doi: 10.1126/science.287.5454.826;pmid: 10657290

41. M. Y. Pecot, V. Malhotra, Golgi membranes remainsegregated from the endoplasmic reticulum during mitosis inmammalian cells. Cell 116, 99–107 (2004). doi: 10.1016/S0092-8674(03)01068-7; pmid: 14718170

42. A. Y. Karpova, D. G. R. Tervo, N. W. Gray, K. Svoboda, Rapidand reversible chemical inactivation of synaptic transmission

in genetically targeted neurons. Neuron 48, 727–735 (2005).doi: 10.1016/j.neuron.2005.11.015; pmid: 16337911

43. R. M. Luik, B. Wang, M. Prakriya, M. M. Wu, R. S. Lewis,Oligomerization of STIM1 couples ER calcium depletion toCRAC channel activation. Nature 454, 538–542 (2008).doi: 10.1038/nature07065; pmid: 18596693

44. J. H. Bayle et al., Rapamycin analogs with differential bindingspecificity permit orthogonal control of protein activity.Chem. Biol. 13, 99–107 (2006). doi: 10.1016/j.chembiol.2005.10.017; pmid: 16426976

45. K. Stankunas et al., Conditional protein alleles using knockinmice and a chemical inducer of dimerization. Mol. Cell 12,1615–1624 (2003). doi: 10.1016/S1097-2765(03)00491-X;pmid: 14690613

46. H. D. Mootz, E. S. Blum, A. B. Tyszkiewicz, T. W. Muir,Conditional protein splicing: A new tool to control proteinstructure and function in vitro and in vivo. J. Am. Chem. Soc.125, 10561–10569 (2003). doi: 10.1021/ja0362813;pmid: 12940738

47. E. R. Ballister, C. Aonbangkhen, A. M. Mayo, M. A. Lampson,D. M. Chenoweth, Localized light-induced proteindimerization in living cells using a photocaged dimerizer.Nat. Commun. 5, 5475 (2014). doi: 10.1038/ncomms6475;pmid: 25400104

48. S. J. Kopytek, R. F. Standaert, J. C. Dyer, J. C. Hu, Chemicallyinduced dimerization of dihydrofolate reductase by ahomobifunctional dimer of methotrexate. Chem. Biol. 7,313–321 (2000). doi: 10.1016/S1074-5521(00)00109-5;pmid: 10801470

49. K. J. Liu, J. R. Arron, K. Stankunas, G. R. Crabtree,M. T. Longaker, Chemical rescue of cleft palate and midlinedefects in conditional GSK-3b mice. Nature 446, 79–82(2007). doi: 10.1038/nature05557; pmid: 17293880

50. K. Nishimura, T. Fukagawa, H. Takisawa, T. Kakimoto,M. Kanemaki, An auxin-based degron system for the rapiddepletion of proteins in nonplant cells. Nat. Methods 6,917–922 (2009). doi: 10.1038/nmeth.1401; pmid: 19915560

51. X. Tan et al., Mechanism of auxin perception by the TIR1ubiquitin ligase. Nature 446, 640–645 (2007). doi: 10.1038/nature05731; pmid: 17410169

52. M. Morawska, H. D. Ulrich, An expanded tool kit for theauxin-inducible degron system in budding yeast. Yeast 30,341–351 (2013). doi: 10.1002/yea.2967; pmid: 23836714

53. M. A. Ko et al., Plk4 haploinsufficiency causes mitoticinfidelity and carcinogenesis. Nat. Genet. 37, 883–888(2005). doi: 10.1038/ng1605; pmid: 16025114

54. C. O. Rosario et al., A novel role for Plk4 in regulating cellspreading and motility. Oncogene 34, 3441–3451 (2015).doi: 10.1038/onc.2014.275; pmid: 25174401

55. A. J. Holland, D. Fachinetti, J. S. Han, D. W. Cleveland,Inducible, reversible system for the rapid and completedegradation of proteins in mammalian cells. Proc. Natl. Acad.Sci. U.S.A. 109, E3350–E3357 (2012). doi: 10.1073/pnas.1216880109; pmid: 23150568

56. B. G. Lambrus et al., p53 protects against genome instabilityfollowing centriole duplication failure. J. Cell Biol. 210, 63–77(2015). doi: 10.1083/jcb.201502089; pmid: 26150389

57. M. Kanke et al., Auxin-inducible protein depletion system infission yeast. BMC Cell Biol. 12, 8 (2011). doi: 10.1186/1471-2121-12-8; pmid: 21314938

58. L. Zhang, J. D. Ward, Z. Cheng, A. F. Dernburg, Theauxin-inducible degradation (AID) system enables versatileconditional protein depletion in C. elegans. Development 142,4374–4384 (2015). doi: 10.1242/dev.129635;pmid: 26552885

59. T. Natsume, T. Kiyomitsu, Y. Saga, M. T. Kanemaki, Rapidprotein depletion in human cells by auxin-inducible degrontagging with short homology donors. Cell Reports 15,210–218 (2016). doi: 10.1016/j.celrep.2016.03.001;pmid: 27052166

60. E. P. Nora et al., Targeted degradation of CTCF decoupleslocal insulation of chromosome domains from genomiccompartmentalization. Cell 169, 930–944.e22 (2017).doi: 10.1016/j.cell.2017.05.004; pmid: 28525758

61. H. D. Ou et al., ChromEMT: Visualizing 3D chromatinstructure and compaction in interphase and mitotic cells.Science 357, eaag0025 (2017). doi: 10.1126/science.aag0025; pmid: 28751582

62. N. A. Hathaway et al., Dynamics and memory ofheterochromatin in living cells. Cell 149, 1447–1460 (2012).doi: 10.1016/j.cell.2012.03.052; pmid: 22704655

63. C. Hodges, G. R. Crabtree, Dynamics of inherently boundedhistone modification domains. Proc. Natl. Acad. Sci. U.S.A.

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 8 of 9

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

109, 13296–13301 (2012). doi: 10.1073/pnas.1211172109;pmid: 22847427

64. J. A. Kennison, J. W. Tamkun, Dosage-dependent modifiers ofpolycomb and antennapedia mutations in Drosophila. Proc.Natl. Acad. Sci. U.S.A. 85, 8136–8140 (1988). doi: 10.1073/pnas.85.21.8136; pmid: 3141923

65. C. Kadoch et al., Dynamics of BAF-Polycomb complexopposition on heterochromatin in normal and oncogenicstates. Nat. Genet. 49, 213–222 (2017). doi: 10.1038/ng.3734; pmid: 27941796

66. B. Z. Stanton et al., Smarca4 ATPase mutations disrupt directeviction of PRC1 from chromatin. Nat. Genet. 49, 282–288(2017). doi: 10.1038/ng.3735; pmid: 27941795

67. S. L. Morgan et al., Manipulation of nuclear architecturethrough CRISPR-mediated chromosomal looping. Nat.Commun. 8, 15993 (2017). doi: 10.1038/ncomms15993;pmid: 28703221

68. E. C. Dykhuizen et al., BAF complexes facilitate decatenationof DNA by topoisomerase IIa. Nature 497, 624–627 (2013).doi: 10.1038/nature12146; pmid: 23698369

69. E. L. Miller et al., TOP2 synergizes with BAF chromatinremodeling for both resolution and formation of facultativeheterochromatin. Nat. Struct. Mol. Biol. 24, 344–352 (2017).doi: 10.1038/nsmb.3384; pmid: 28250416

70. S. Gruber et al., Evidence that loading of cohesin ontochromosomes involves opening of its SMChinge.Cell 127, 523–537(2006). doi: 10.1016/j.cell.2006.08.048; pmid: 17081975

71. J. E. Gestwicki, G. R. Crabtree, I. A. Graef, Harnessingchaperones to generate small-molecule inhibitors of amyloidb aggregation. Science 306, 865–869 (2004). doi: 10.1126/science.1101262; pmid: 15514157

72. P. S. Marinec et al., FK506-binding protein (FKBP) partitionsa modified HIV protease inhibitor into blood cells and prolongsits lifetime in vivo. Proc. Natl. Acad. Sci. U.S.A. 106, 1336–1341(2009). doi: 10.1073/pnas.0805375106; pmid: 19164520

73. S. C. Penchala et al., AG10 inhibits amyloidogenesis andcellular toxicity of the familial amyloid cardiomyopathy-associated V122I transthyretin. Proc. Natl. Acad. Sci. U.S.A.110, 9992–9997 (2013). doi: 10.1073/pnas.1300761110;pmid: 23716704

74. T. Ito et al., Identification of a primary target of thalidomideteratogenicity. Science 327, 1345–1350 (2010). doi: 10.1126/science.1177319; pmid: 20223979

75. E. S. Fischer et al., Structure of the DDB1-CRBN E3 ubiquitinligase in complex with thalidomide. Nature 512, 49–53(2014). doi: 10.1038/nature13527; pmid: 25043012

76. P. Filippakopoulos et al., Selective inhibition of BETbromodomains. Nature 468, 1067–1073 (2010).doi: 10.1038/nature09504; pmid: 20871596

77. G. E. Winter et al., Phthalimide conjugation as a strategy forin vivo target protein degradation. Science 348, 1376–1381(2015). doi: 10.1126/science.aab1433; pmid: 25999370

78. J. Lu et al., Hijacking the E3 ubiquitin ligase cereblon toefficiently target BRD4. Chem. Biol. 22, 755–763 (2015).doi: 10.1016/j.chembiol.2015.05.009; pmid: 26051217

79. D. P. Bondeson et al., Catalytic in vivo protein knockdown bysmall-molecule PROTACs. Nat. Chem. Biol. 11, 611–617(2015). doi: 10.1038/nchembio.1858; pmid: 26075522

80. M. A. Erb et al., Transcription control by the ENL YEATSdomain in acute leukaemia. Nature 543, 270–274 (2017).doi: 10.1038/nature21688; pmid: 28241139

81. J. E. Delmore et al., BET bromodomain inhibition as atherapeutic strategy to target c-Myc. Cell 146, 904–917(2011). doi: 10.1016/j.cell.2011.08.017; pmid: 21889194

82. C. J. Ott et al., BET bromodomain inhibition targets bothc-Myc and IL7R in high-risk acute lymphoblastic leukemia.Blood 120, 2843–2852 (2012). doi: 10.1182/blood-2012-02-413021; pmid: 22904298

83. D. L. Buckley et al., Targeting the von Hippel–Lindau E3ubiquitin ligase using small molecules to disrupt the VHL/HIF-1a interaction. J. Am. Chem. Soc. 134, 4465–4468(2012). doi: 10.1021/ja209924v; pmid: 22369643

84. D. L. Buckley et al., HaloPROTACS: Use of small moleculePROTACs to induce degradation of HaloTag fusion proteins.ACS Chem. Biol. 10, 1831–1837 (2015). doi: 10.1021/acschembio.5b00442; pmid: 26070106

85. A. C. Lai et al., Modular PROTAC design for the degradationof oncogenic BCR-ABL. Angew. Chem. 55, 807–810 (2016).doi: 10.1002/anie.201507634; pmid: 26593377

86. A. Hamilton et al., Chronic myeloid leukemia stem cells arenot dependent on Bcr-Abl kinase activity for their survival.Blood 119, 1501–1510 (2012). doi: 10.1182/blood-2010-12-326843; pmid: 22184410

87. V. M. Rivera et al., Long-term pharmacologically regulatedexpression of erythropoietin in primates followingAAV-mediated gene transfer. Blood 105, 1424–1430(2005). doi: 10.1182/blood-2004-06-2501;pmid: 15507527

88. V. M. Rivera et al., A humanized system for pharmacologiccontrol of gene expression. Nat. Med. 2, 1028–1032 (1996).doi: 10.1038/nm0996-1028; pmid: 8782462

89. X. Ye et al., Regulated delivery of therapeutic proteins after invivo somatic cell gene transfer. Science 283, 88–91 (1999).doi: 10.1126/science.283.5398.88; pmid: 9872748

90. D. M. Spencer et al., Functional analysis of Fas signaling invivo using synthetic inducers of dimerization. Curr. Biol. 6,839–847 (1996). doi: 10.1016/S0960-9822(02)00607-3;pmid: 8805308

91. R. A. MacCorkle, K. W. Freeman, D. M. Spencer, Syntheticactivation of caspases: Artificial death switches. Proc. Natl.Acad. Sci. U.S.A. 95, 3655–3660 (1998). doi: 10.1073/pnas.95.7.3655; pmid: 9520421

92. L. Fan, K. W. Freeman, T. Khan, E. Pham, D. M. Spencer,Improved artificial death switches based on caspases andFADD. Hum. Gene Ther. 10, 2273–2285 (1999). doi: 10.1089/10430349950016924; pmid: 10515447

93. U. B. Pajvani et al., Fat apoptosis through targeted activationof caspase 8: A new mouse model of inducible and reversiblelipoatrophy. Nat. Med. 11, 797–803 (2005). doi: 10.1038/nm1262; pmid: 15965483

94. D. J. Baker et al., Clearance of p16Ink4a-positive senescentcells delays ageing-associated disorders. Nature 479,232–236 (2011). doi: 10.1038/nature10600; pmid: 22048312

95. D. J. Baker et al., Naturally occurring p16Ink4a-positive cellsshorten healthy lifespan. Nature 530, 184–189 (2016).doi: 10.1038/nature16932; pmid: 26840489

96. A. Di Stasi et al., Inducible apoptosis as a safety switch foradoptive cell therapy. N. Engl. J. Med. 365, 1673–1683 (2011).doi: 10.1056/NEJMoa1106152; pmid: 22047558

97. T. Clackson et al., Redesigning an FKBP-ligand interface togenerate chemical dimerizers with novel specificity.Proc. Natl. Acad. Sci. U.S.A. 95, 10437–10442 (1998).doi: 10.1073/pnas.95.18.10437; pmid: 9724721

98. X. Zhou et al., Inducible caspase-9 suicide genecontrols adverse effects from alloreplete T cells afterhaploidentical stem cell transplantation. Blood 125,4103–4113 (2015). doi: 10.1182/blood-2015-02-628354;pmid: 25977584

99. X. Zhou et al., Long-term outcome after haploidentical stemcell transplant and infusion of T cells expressing the induciblecaspase 9 safety transgene. Blood 123, 3895–3905 (2014).doi: 10.1182/blood-2014-01-551671; pmid: 24753538

100. J. N. Kochenderfer, S. A. Rosenberg, Treating B-cell cancerwith T cells expressing anti-CD19 chimeric antigen receptors.Nat. Rev. Clin. Oncol. 10, 267–276 (2013). doi: 10.1038/nrclinonc.2013.46; pmid: 23546520

101. C. R. Cruz et al., Infusion of donor-derived CD19-redirectedvirus-specific T cells for B-cell malignancies relapsed afterallogeneic stem cell transplant: A phase 1 study. Blood 122,2965–2973 (2013). doi: 10.1182/blood-2013-06-506741;pmid: 24030379

102. I. Diaconu et al., Inducible caspase-9 selectively modulatesthe toxicities of CD19-specific chimeric antigen receptor-modified T cells. Mol. Ther. 25, 580–592 (2017). doi: 10.1016/j.ymthe.2017.01.011; pmid: 28187946

ACKNOWLEDGMENTS

We apologize to the many people who have made importantcontributions to this field whom we were unable to acknowledgebecause of space constraints. We wish to honor the lastingmemory of Joseph P. Calarco, who was a great friend and incisivecolleague. E.J.C. was supported by an NSF Graduate ResearchFellowship and the Ruth L. Kirschstein National Research ServiceAward (F31 CA203228-02). G.R.C. active-research funding includessupport from the Howard Hughes Medical Institute, NIH5R01CA163915-04, the NIH Javits Neuroscience Investigator AwardR37 NS046789-12, the Congressionally Directed Medical ResearchProgram Breast Cancer Research Breakthrough Award, and theSimons Foundation Autism Research Initiative. The authors declareno competing interests. G.R.C. is an inventor on several patentsand patent applications related to chemical inducers of proximityand the uses thereof, filed between 1997 and 2017. These includepatent applications WO2017074943A1, WO2013188406A1,US20130158098A1, and WO2002024957A1 submitted by TheBoard of Trustees of Leland Stanford Junior University,applications US20110160246A1 and US20110003385A1 submittedby G.R.C., patent application US20020173474A1 submitted by thePresident and Fellows of Harvard College. Patents US8084596B2,US6984635B1, and US6063625A are held by The Board ofTrustees of Leland Stanford Junior University and the Presidentand Fellows of Harvard College.

10.1126/science.aao5902

Stanton et al., Science 359, eaao5902 (2018) 9 March 2018 9 of 9

RESEARCH | REVIEWon O

ctober 1, 2020

http://science.sciencemag.org/

Dow

nloaded from

Chemically induced proximity in biology and medicineBenjamin Z. Stanton, Emma J. Chory and Gerald R. Crabtree

DOI: 10.1126/science.aao5902 (6380), eaao5902.359Science 

, this issue p. eaao5902Sciencerange of advances and speculate on future applications of this fundamental approach.

review the wideet al.This ''induced proximity'' can also be applied to the development of new therapeutics. Stanton synthetic biology, signal transduction, transcription, protein degradation, epigenetic memory, and chromatin dynamics.membrane-permeable small molecules that reversibly regulate proximity has enabled advances in fields such as

The physical distance, or proximity, between molecules often directs biological events. The development ofRegulating molecule proximity

ARTICLE TOOLS http://science.sciencemag.org/content/359/6380/eaao5902

REFERENCES

http://science.sciencemag.org/content/359/6380/eaao5902#BIBLThis article cites 102 articles, 34 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on October 1, 2020

http://science.sciencem

ag.org/D

ownloaded from